Abstract Obesity prevention has become a national health imperative, and may be exceptionally important yet particularly difficult during adolescence. Adolescents develop enduring lifestyle patterns in their dietary choices and activity levels that can set the stage for life-long obesity, so it is essential that the field identify novel modifiable factors that impact the lifestyle patterns of adolescents. This study tests a promising new target for prevention efforts. Our group has collected pilot data that builds upon prior correlational findings in children and experimental findings in adults to suggest that sleep restriction (SR) causes adolescents to eat more calories without increasing their physical activity. If these pilot findings are confirmed in a larger, more definitive study, obesity prevention efforts could be augmented by strategies to alleviate SR, which is known to be extremely common among adolescents. Further, our pilot findings with adolescents extend prior adult research to suggest that the mechanism for this effect is a particular increase in the reward value of sweet/ dessert foods, causing them to be eaten in greater amounts, to be perceived more positively, and to elicit more activation of reward-related brain regions on fMRI during SR. Confirmation of these pilot findings will inform the complex biobehavioral foundations of obesity, and identify secondary points of intervention for occasions when an adolescent's sleep is inadequate due to unusual circumstances or to an intractable sleep disorder. However, before such benefits can be realized, the field must overcome significant gaps in our knowledge base. Our pilot studies yielded intriguing findings, but were too small and methodologically limited to be definitive. Other pediatric findings have come almost exclusively from correlational studies, greatly limiting causal inferences. Adult SR findings also are difficult to apply to adolescents because of multiple methodological and developmental factors. We propose to overcome these limitations with a well-powered mechanistic study that balances experimental rigor with real-world relevance during the high-risk period of adolescence. Healthy, non- obese 14-17 year-olds will undergo a randomized cross-over trial of an experimental sleep manipulation across 5-night periods (simulating a school week), comparing nightly sleep of (a) ~9 hr, the recommendation for healthy sleep vs. (b) ~6.3 hr, which many teens experience on school nights (SR). Lab assessments of teens' perception of, and brain responses to, specific foods will be paired with measures of real-world dietary and activity patterns to definitively test (a) the effect of SR on the caloric intake and activity of healthy adolescents, (b) the effect of SR on the rewarding quality of calorie-dense foods, and (c) the link between changes in caloric intake and the rewarding quality of food. Analyses will further explore whether risk factors for obesity identify which adolescents are especially vulnerable to the effect of SR on caloric intake. Findings will have direct implications for improving obesity prevention and for discoveries into obesity's complex biobehavioral mechanisms.